property-management

Implement one task end-to-end for the PPT project — pick the right per-stack specialist, code, verify per-stack, push a draft PR against `dev`.

Project-management & delivery analysis for the PPT research routine (Phase 1.6). Runs an always-on Scrum Master synthesis plus role-based deep analysis (rotating one role/day by default; all 8 on `full`; a specific role on `pm:<role>`). Reads sprint-status + repo activity + research backlog, spawns role subagents, and writes delivery artifacts under .research/management/. Use from routine Phase 1.6, or standalone for a delivery snapshot.

Post-merge code review with issue creation. Reviews merged PRs in a time window, spots better approaches / missed edge cases / security holes / perf issues / test gaps, and opens GitHub issues with concrete improvement proposals.

Deterministic goal/convergence checks for the PPT dispatcher — coverage referential integrity, coverage-progress monotonicity, and buffer bounds. Run in dispatcher Phase 6 and CI. Use when adding/auditing the dispatcher's goal-verification layer.

Open a PR in this project's style — title, body template, IG3 evidence, CI surface, draft handling.

Land a green, approved PR. Verify preconditions (CI green, approved, no unresolved threads, not draft), auto-resolve mechanical merge conflicts against the base branch (sqlx offline data, Cargo.lock, generated openapi/api-client, lockfiles), then `gh pr merge --squash --auto`. Stops and surfaces if real code conflicts or stale CI.

Close out a PR's post-push loop. Two modes — (a) `dispatcher` — read the dispatcher's reviewer verdict in assignments.json and respawn the original implementer on `verdict=changes` (called by ppt-research-dispatcher after Phase 5 returns changes); (b) `manual` — implementer-driven CI + review-comment loop on the current branch (called right after ppt-pr-create or by a human). Triggers on "follow up on the PR", "fix the CI", "address review comments", "respawn implementer".

Orchestrates a multi-expert development team for PPT (Property Management) code review and fixing. Use this skill whenever you want a thorough review before merging, releasing, or completing an epic. Triggers on: "run the dev team", "10-pass review", "full team review", "expert team review", "review with the team", "team review before PR", "run all experts on this". The team has 8 roles — Rust, Kotlin, Frontend, Completeness, Owner, Scrum Master, Planner, Tester — runs 10 structured review passes, then coordinates all fixes. One Rust compile at a time, always.

Rotating expert code review for the PPT research routine. Each run reviews ONE scope segment (from 8 rotating segments covering the full codebase) using static analysis only — no compilation, no test execution. Findings are emitted as `code-review-finding` signals for the research backlog. Invoke from Phase 1.5 of the research routine. Segment selection is automatic: churn hotspots from Phase 1 get priority, otherwise picks the oldest-unreviewed segment. Use this skill whenever the research routine runs its daily Phase 1.5 code review slice.

Author + validate screen-map docs, keep them in sync with routes, and run screen smoke checks via Chrome MCP / Playwright.

Pick the next plan to ship from .research/backlog.json — highest-score ready row, deterministic tiebreak.

Interact with the research routine itself — fire `deep` or `reset` runs from a Claude Code session without crafting curl by hand.

Use the ppt-bridge MCP to run dev-stack, DB queries, and tests against a remote host from a cloud routine.

Apply / author SQLx migrations for the PPT databases — and document the missing-seed-recipe gap.

Manage worktree deployments to *.dev.ppt.rlt.sk and *.dev.rlt.sk via the ppt-deploy server. Use when the user wants to spin up, list, status, or close a worktree dev environment, or to deploy/promote staging or prod. Triggers on phrases like "open worktree", "deploy this branch", "promote v1.2.3", "close worktree".

Bring up the local PPT dev stack declaratively via `stack up pm-local`, with a Docker Compose fallback.

Navigate the React/pnpm frontend workspace — apps, shared packages, generated API client, Biome.

Kotlin Multiplatform app under mobile-native/ — Gradle workflow, modules, ADB requirement.

Drive a plan from .research/plans/ through implementation, PR, and post-merge archive.

Navigate the backend Cargo workspace — crate layout, per-crate dev loop, sqlx workflow, seed data.

Pick the smallest correct test command for the change at hand — backend, frontend, integration, or mobile.

Author TypeSpec at docs/api/typespec/ to drive the OpenAPI spec and the generated Rust + TS clients.

Drive a real Windows Chrome/Edge/Firefox window from a WSL agent via Windows-side Node.js + Playwright. Use when the task needs a visible browser window, real clicking/typing, codegen, or headed debugging — not headless Linux. Triggers on phrases like "open the browser", "click through this page", "playwright codegen", "headed test", "visible browser", "real browser".

Detect drift between code (sitemap, use-cases, epics) and screen-maps. Reports unmapped sitemap entries, unknown endpoints, missing components/use-cases/epics, and orphan screens. Interactive prompt to apply patches. Use when the user runs `/screens update` or asks to "check screen-map drift", "screens stale", "what's out of date".

Query the screen-map tree by frontmatter filter (e.g. "product:ppt,implementations.ppt-web.redesignStatus:in-progress"). Output as table, JSON, or markdown. Use when user runs `/screens query` or asks "which screens are X", "list all Y screens", "find screens with Z".

Generate Mermaid diagrams from the screen-map tree — site graph (screens × related-screen edges), endpoint matrix (screens × endpoints), status dashboard (per-platform pie charts of build/redesign/api status). Output to docs/screens/_diagrams/. Use when user runs `/screens render` or asks for "screen-map diagrams", "visualize screens", "status dashboard".

Load focused context for a single screen-map (parent + children + related + sitemap entries + recent agent log + optional Playwright screenshot). Use when the user runs `/screens edit <id>` or asks to "load context for screen X", "show me ppt/foo", "what's the current state of <screen>".

Bootstrap or extend docs/screens/ for one product by scanning sitemap, use-cases, epics, an optional design ZIP, and any user-provided candidate names; then walks the user through interactive grouping in chat before bulk-writing the markdown files. Use when the user asks to "init screens", "bootstrap screen-map", "create screen maps for ppt|reality", or runs `/screens init`.

Spawn the local Visual Review server (Hono + Preact SPA) on 127.0.0.1, open the browser, and walk every screen-map for a product with OK / Note checkboxes and a per-screen Save & Next button. Notes persist directly into Agent Log + Notes > Specific. Use when the user runs `/screens review` or asks to "review screens", "walk the screen-map", "do a redesign review".

Validate every screen-map under docs/screens/ against frontmatter schema, @ppt/sitemap IDs, related-screen IDs, and diagram refs. Use when the user runs `/screens validate` or asks to check screen-map consistency. Triggers also include "validate screens", "screen-map drift", "is the screen-map clean".